Image Matting
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 1
- 평가표
- 8
- 보존된 평가 기록
- 45
Image Matting is the process of accurately estimating the foreground object in images and videos. It is a very important technique in image and video editing applications, particularly in film production for creating visual effects. In case of image segmentation, we segment the image into foreground and background by labeling the pixels. Image segmentation generates a binary image, in which a pixel either belongs to foreground or background. However, Image Matting is different from the image segmentation, wherein some pixels may belong to foreground as well as background, such pixels are called partial or mixed pixels. In order to fully separate the foreground from the background in an image, accurate estimation of the alpha values for partial or mixed pixels is necessary.
Source: Automatic Trimap Generation for Image Matting
Image Source: Real-Time High-Resolution Background Matting
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
MP-Mat: A 3D-and-Instance-Aware Human Matting and Editing Framework with Multiplane Representation
MaSS13K: A Matting-level Semantic Segmentation Benchmark
VRMDiff: Text-Guided Video Referring Matting Generation of Diffusion
Path-Adaptive Matting for Efficient Inference Under Various Computational Cost Constraints
Object-Aware Video Matting with Cross-Frame Guidance
Enhancing Image Matting in Real-World Scenes with Mask-Guided Iterative Refinement
Efficient Portrait Matte Creation With Layer Diffusion and Connectivity Priors
MatAnyone: Stable Video Matting with Consistent Memory Propagation
BEN: Using Confidence-Guided Matting for Dichotomous Image Segmentation
ReFlow6D: Refraction-Guided Transparent Object 6D Pose Estimation via Intermediate Representation Learning
ILDiff: Generate Transparent Animated Stickers by Implicit Layout Distillation
Memory Efficient Matting with Adaptive Token Routing
Morpho-Aware Global Attention for Image Matting
ZIM: Zero-Shot Image Matting for Anything
Automating Video Thumbnails Selection and Generation with Multimodal and Multistage Analysis
DFIMat: Decoupled Flexible Interactive Matting in Multi-Person Scenarios
Towards Natural Image Matting in the Wild via Real-Scenario Prior
Alfie: Democratising RGBA Image Generation With No $$$
Training Matting Models without Alpha Labels
Matting by Generation
FADE: A Task-Agnostic Upsampling Operator for Encoder-Decoder Architectures
Boosting General Trimap-free Matting in the Real-World Image
MaGGIe: Masked Guided Gradual Human Instance Matting
Towards Label-Efficient Human Matting: A Simple Baseline for Weakly Semi-Supervised Trimap-Free Human Matting
Learning Trimaps via Clicks for Image Matting
Learning Multiple Representations with Inconsistency-Guided Detail Regularization for Mask-Guided Matting
In-Context Matting
What Matters When Repurposing Diffusion Models for General Dense Perception Tasks?
DiffuMatting: Synthesizing Arbitrary Objects with Matting-level Annotation
End-to-End Human Instance Matting
Dual-Context Aggregation for Universal Image Matting
DART: Depth-Enhanced Accurate and Real-Time Background Matting
POBEVM: Real-time Video Matting via Progressively Optimize the Target Body and Edge
Lightweight high-resolution Subject Matting in the Real World
Diffusion for Natural Image Matting
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.